Fusion program in doldrums

When an experimental fusion device passed an early start-up test at Princeton University Christmas Eve, it was hailed as a milestone in the US effort to control the ''fire'' of the stars. But that momentary elation hasn't eased experts' concern for the future of this highly successful program.

A report of a fusion workshop, distributed recently by the National Academies of Sciences and of Engineering, praises the program's scientific progress. Yet it warns that engineering knowledge needed to turn this basic understanding into an operating fusion power reactor lags.

This reflects Reagan administration cutbacks in what, by congressional mandate, was to have been a fast-paced fusion engineering development effort.

The Magnetic Fusion Engineering Act of 1980 envisioned sharp budget increases. It called for an experimental device to test materials and other engineering aspects of fusion reactors. Construction was to start by 1987. It urged, hopefully, that an experimental fusion power reactor be operating by the year 2000.

Instead, the budget has been held down strongly. The engineering test device has been canceled with a promise of building it in the 1990s. And the prospect of a prototype power reactor has again been postponed to sometime in the next century. This has led Alvin W. Trivelpiece, director of the Department of Energy's Office of Energy Research, to observe that ''the economic conditions are such that, regardless of the merits of the activity, substantial new growth in fusion is unlikely.''

Magnetic fusion - the most advanced type of fusion control - refers to devices in which magnetic fields contain the hot (100 million degrees C.) gases in which fusion takes place. The academy report says the US has a magnetic fusion program based on ''mature science'' but on engineering knowledge that is ''less mature.'' It warns: ''. . . those concepts based on the best physical understanding lead to reactor embodiments with engineering complexities that will be very difficult to resolve. Conversely, those concepts that apparently offer attractive reactors are ones lacking a mature basis in physics.''

In short, the report makes a strong case for developing the needed engineering knowledge as rapidly as possible. This is unlikely with the present budget trend.

The Magnetic Fusion Advisory Committee (MFAC) has reached similarly disturbing conclusions. Trivelpiece assembled the MFAC hastily last year to assess the program. He asked it to consider the outlook under three different constant annual funding levels of $400 million, $500 million, and $600 million. The fiscal 1983 magnetic fusion budget request was $444 million.

A recently distributed summary of the MFAC report warns that meaningful progress can be made only at the $600 million funding level. At $500 million, there would be delays in developing engineering knowledge ''marked by a substantial increase in technical risk in establishing the necessary data base for an early . . . decision (on building an engineering test reactor.) The report adds that the $400 million funding level would have ''a major negative impact.''

The academy study sums up the status of US fusion research as having to face ''the twin challenge for sustained and closely coupled programs of scientific and engineering development.'' The Reagan administration has yet to face this challenge squarely. Here comes the 'supertomato'

Genetic engineering now promises a heartier, meatier, faster-growing tomato.

Heinz USA has teamed up with ARCO Solar Industries, a division of Atlantic Richfield Company, to produce such a supertomato by trying to implant the genes of certain wild tomatoes into the chromosomes of a commercial variety. The aim is to mate some of the best traits, such as cold resistance, of the wild types with the juiciness and other marketable qualities of cultivated plants. The Heinz-ARCO venture follows one launched last March between the Campbell Soup Company and DNA Plant Technology Corporation, a New Jersey bioengineering firm.

In both experiments, the thrust is to produce a better ''processing'' tomato for use in ketchup, soup, paste, and sauces. ''We want to take that backyard garden flavor and put it in a processing tomato,'' says David Evans of DNA Plant Technology.

A chief goal will be to cut the water content. The average commercial tomato is roughly 95 percent fluids. Any water content reduction would cut transportation and processing costs. Eventually, researchers say, they think genetic engineering will improve taste and growing habits of all tomatoes. ''Once the breakthroughs start occurring, the benefits will flow to a lot of other crops,'' says Walter Tusinski, an ARCO bioengineering expert. Chemicals from the clouds

Mountain forests extract much water from clouds. Conifer needles are especially well adapted to gather such moisture. But what of chemicals in that water?

Gary M. Lovett of the US Oak Ridge National Laboratory and William A. Reiners and Richard K. Olson of Dartmouth College have data which they say suggest ''extremely high rates of . . . deposition of both beneficial substances, such as inorganic nitrogen, and potentially detrimental ones, such as mineral acids and trace metals.'' Reporting this in Science, they add, ''As a result, subalpine ecosystems may show the most exaggerated accumulations and effects of these substances in the Northeastern United States.'' How cold the ice age?

Peak cooling in the last ice age was 7 to 8 degrees C. below today's normal, with as much as 2 meters extra precipitation during full glacial times in California.

David P. Adams of the US Geological Survey and G. James West of the US Bureau of Reclamation have estimated these conditions by studying fossil pollen from Clear Lake in California. They explain in Science how they calibrated the pollen thermometer using modern data.

Modern pollen was taken from six lake and marsh sites in Northern California. Variations in the ratio of oak to oak-plus-pine pollen were than translated into temperature and precipitation changes. This gave them a relationship which enabled them to use fossil pollen as a thermometer and precipitation guage.

They note that the precipitation estimate probably is less reliable than that for temperature. And they admit that the estimates ''are crude.'' But, they add, they ''provide a better measure of past temperature and precipitation changes in the northern Coast Ranges of California than has heretofore been available.''